From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from smtp10.smtpout.orange.fr ([80.12.242.132] helo=smtp.smtpout.orange.fr) by bombadil.infradead.org with esmtps (Exim 4.80.1 #2 (Red Hat Linux)) id 1ZrXrY-0004YY-Ln for linux-mtd@lists.infradead.org; Wed, 28 Oct 2015 21:02:29 +0000 From: Robert Jarzmik To: Brian Norris Cc: Boris Brezillon , Marek Vasut , linux-mtd@lists.infradead.org, linux-kernel@vger.kernel.org, Ezequiel Garcia , Scott Wood , Josh Wu , Kyungmin Park , Han Xu , Huang Shijie Subject: Re: [PATCH 1/5] mtd: ofpart: grab device tree node directly from master device node References: <1445913070-17950-1-git-send-email-computersforpeace@gmail.com> <20151027175446.GT13239@google.com> <20151028085813.4b0b3ac8@bbrezillon> <201510281711.14196.marex@denx.de> <20151028173215.0c1c4e30@bbrezillon> <20151028171430.GC13239@google.com> Date: Wed, 28 Oct 2015 21:55:24 +0100 In-Reply-To: <20151028171430.GC13239@google.com> (Brian Norris's message of "Wed, 28 Oct 2015 10:14:30 -0700") Message-ID: <87eggek91f.fsf@belgarion.home> MIME-Version: 1.0 Content-Type: text/plain List-Id: Linux MTD discussion mailing list List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Brian Norris writes: >> > >> > Do some sorts of chipselects come into play here ? Ie. you can have one master >> > with multiple NAND chips connected to it. >> >> Most NAND controllers support interacting with several chips (or >> dies in case your chip embeds several NAND dies), but I keep thinking >> each physical chip should have its own instance of nand_chip + mtd_info. >> If you want to have a single mtd device aggregating several chips you >> can use mtdconcat. >> >> This leaves the multi-dies chip case, and IHMO we should represent those >> chips as a single entity, and I guess that's the purpose of the >> ->numchips field in nand_chip (if your chip embeds 2 dies with 2 CS >> lines, then ->numchips should be 2). > Yes, I think that's some of the intention there. And so even in that > case, a multi-die chip gets represented as a single struct nand_chip. Isn't there the case of a single NAND controller with 2 identical chips, each a 8 bit NAND chip, and the controller aggregating them to offer the OS a single 16-bit NAND chip ? In this case, the controller (pxa3xx is a good example) will be programmed to handle both chips at the same time, and calculate CRC on both chips, etc ... I hope the assertion "physical chip should have its own instance of nand_chip + mtd_info" does take into account this example. I don't know if there is actually any user of this for either pxa3xx or another controller, nor if there is any value in this. Cheers. -- Robert